International Journal on Recent and Innovation Trends in Computing and Communication

Volume: 2 Issue: 5

ISSN: 2321-8169
1081 1085

_______________________________________________________________________________________________

An Automated Skin Lesion Diagnosis by using Image Processing Techniques

Sujaya Saha

Dr. Rajat Gupta

M.Tech Final Year Student, Deptt. Of ECE, MMU,

Mullana, Ambala
sujaya5711@gmail.com

Professor, Deptt. Of ECE, MMEC,

Mullana, Ambala
guptarajat46@gmail.com

Abstract The application of image processing for diagnostics purpose is a non-invasive technique. At present there is a great interest in the
prospects of automatic image analysis method for image processing, which provides significant information about a skin lesion, also can be
more applicable for the clinical purpose, and as an early warning tool for the detection purpose.. In order to accomplish an efficient way to
identify skin cancer at an early stage without performing any unnecessary skin biopsies, digital images of skin lesions have been investigated. To
complete this goal, feature extraction is considered as an essential-weapon to analyze an image properly. In this paper, different digital lesion
images have been analyzed based on unsupervised image acquisition, pre-processing, and image segmentation techniques. Then the Feature
extraction techniques are applied on these segmented images. After this, a graphical user interface has been designed for the lesion probability
detection and then a comprehensive discussion has been explored based on the obtained results.

Keywords Skin Cancer, Image Acquisition, Image Segmentation, Feature Extraction, Graphical User Interface, Skin Cancer
Detection.
__________________________________________________*****_________________________________________________
I. INTRODUCTION
Cancer begins when cells in the part of the body starts to
grow out of control. A lesion means possibly abnormal
change or difference in a tissue or structure such as the skin.
Cancer is an uncontrolled growth of abnormal cells. The
skin cancer is the uncontrolled growth of skin cells in the
body. It develops when an unrepaired DNA damage to the
skin cells and mostly caused by ultraviolet radiation of the
sun or tanning beds, trigger mutations (genetic defects)
which leads to the skin cells multiply rapidly and malignant
tumours form. Some skin cancer can spread and cause
damage in the nearby tissue cells [1]. Also, in some cases,
skin cancer can be on vital organs. Sun is the most common
cause of skin cancer. But it fully does not explain that skin
cancer usually develop on the skin exposed to sunlight. Also
it can be exposed to environmental threats, radiation
analysis, and even inheritance could play a role. Although
anyone can get skin cancer, the risk is greatest for people
having bright skin or bright eyes, a wealth of large, irregular
shaped moles, a genetic family history of skin cancer, an
excessive sunlight or sunburn, lived in large or with yearround sunshine, received radiation medication [3].
The sign of skin cancer often starts as the change of color in
the skin. They are usually mixed color (pink, red, and
brown). There are three types of skin cancer that occurred.
They are- Basel cell cancer, squamous cell carcinoma and
malignant melanoma tumour. The first two does not spread
quickly, but the third one spreads quickly. Melanoma is
much less common than basal cell and squamous cell skin
cancer, but it is far more dangerous than the other two types.
However, it is much more dangerous if it is not found early.

It causes the majority (75%) of deaths related to the skin

cancer [4]. Worldwide, doctors investigate about 160,000
new cases of melanoma per year. It is mostly common in
women than in men. In women, the most common part that
the cancer occurs is the legs and in case of men the most
common part is on the back. .It is specifically common
among Caucasians, and especially north Europeans those
who lives in sunny climates. There are high rates of
occurrences in Australia, North and South America, and
North Europe, also with a disordered decrease in southern
Italy and Sicily. This geographic pattern mirrors the primary
cause, ultraviolet light (UV) exposure crossed with the
amount of skin pigmentation in the population [5]. There
can be new growths or precancerous lesions that are not
cancer but could become cancer over time. An estimated
40% to 50% of fair-skinned people who live to be 65 will
develop at least one skin cancer. Skin cancer can be cured if
it's found and treated early. The possible signs of skin cancer
include a change in the appearance of a mole or pigmented
area. If a mole changes its size, shape, or color, and has
irregular edges, and it is having more than one color, also it
is asymmetrical, or itches, oozes, or bleeds then there is
possibility of skin cancer .The treatment includes surgical
removal of the tumour. If melanoma is found early, while it
is still small and thin, then the chance of cure is high. The
main design issues for the proper characterization of skin
lesions of malignant tumour is the image acquisition, the
image pre-processing and analysis, the image segmentation ,
the feature extraction, and the detection [6].
II. PROPOSED WORK
1081

IJRITCC | May 2014, Available @ http://www.ijritcc.org

_______________________________________________________________________________________

International Journal on Recent and Innovation Trends in Computing and Communication

Volume: 2 Issue: 5

ISSN: 2321-8169
1081 1085

_______________________________________________________________________________________________
At first an image is acquired with a digital camera under
consistent lighting. The proper interpretation of these
dermoscopic images leads to increased clinical diagnostic
accuracy. Most Automated Skin Lesion Diagnosis methods
adopt the standard computer-aided diagnosis (CAD) pipeline
which is illustrated in Fig. 1 below, and it consists of five
general stages. After the image is acquired, it contains many
artifacts such as hair and oil bubbles which could bias

Image
Acquisition

Image Preprocessing

downstream processes are identified. Next, the lesion is

segmented from the surrounding healthy skin. After
segmentation, discriminative features are extracted from the
lesion. Features which are usually extracted are border,
colour, entropy, compactness, radial variance of the mask,
coarseness. Finally; by extracting these features the
detection is done which finally shows the risk probability of
the lesion which is present in the image.

Image
Segmentation

Feature
Extraction

Detection

Figure 1: Implementation of CAD pipeline for Automated Skin Lesion Diagnosis

III. RESULT AND DISCUSSION
The CAD system performance detection is based on the
Feature Extracted from the image analyzed. The
experimental results are conducted using Matlab7.12.0. For
this experiment image database is used from practical lab.

An Input Image is converted into grey Scale.

Then, after displaying the input image the pre-processing
stage is being carried out. Image Pre-Processing is a
technique where complete analysis of the pigmented skin
lesion is done.

Figure 3.1: Input image

This is an input image taken for the diagnosis purpose.
At first an input image is read and then displayed. The figure
is shown below.

Figure 3.3: Image Pre-processing by Median Filter,

Gaussian Filter and Thresholding.
Here, by median filtering the noise reduction is done i.e. the
noise which is present is removed. Such noise reduction in
pre-processing step is done to improve the results of later
processing. And also it preserves edges while removing
noise. Then, another filter is applied for pre-processing. And
that is Gaussian Filter. By applying this filter constant noise
level in dark areas of the image is removed. But as the
Gaussian filter is implemented it creates blur. So as such it
also gets difficult to analysis a lesion. So due to this effect,
this image is further used to create binary image in the form
of 0 and 1. And so thresholding process is carried out.

Figure 3.2: Input Image and Input image (Grey Scale)

1082
IJRITCC | May 2014, Available @ http://www.ijritcc.org

_______________________________________________________________________________________

International Journal on Recent and Innovation Trends in Computing and Communication

Volume: 2 Issue: 5

ISSN: 2321-8169
1081 1085

_______________________________________________________________________________________________
After that Segmentation part is further processed. The
figures are shown below.

Figure 3.6: Traced Border

Here, the border has been shown with x and y plots.
Figure 3.4: Image Segmentation by Otsu Method and Color
Image Segmentation Method.
Image Segmentation refers to partitioning of an image into
region. Basically the main goal for segmentation is to
simplify and change the representation of an image into
more meaningful and easier way to analysis. By applying
Otsu Segmentation Method it measure the spreading of the
pixel levels each side of the threshold, i.e. the pixels that
either falls in foreground or background. The aim is to find
the threshold value where the sum of foreground and
background spreads is at its minimum. The next
segmentation method used is Color Image Segmentation.
Here three clusters are shown for the proper extraction
purpose. As from the three clusters, Cluster 2 gives proper
boundary plot, so this figure is referred for further process in
the feature extraction stage. Next, comes Feature Extraction
part. Here all the relative features are extracted from the
image.

Figure 3.7: New matrix boundary points and Inner

boundaries
Here, new matrix boundaries are created to get rid of the
other distortions outside the boundary. And the inner
boundaries are filled to detect the lesion mask properly.
Then, the Geometrical Centre is extracted to determine the
coarseness, color entropy, color variance, compactness and
GLRLM (grey level run length method).

Figure 3.5: Border Selection and Traced

Here from the above figure, it can be seen that the border
selection is done from the segmented color cluster 2 and
after that a traced mask is drawn.

Figure 3.8: Geometrical Centre (color)

1083
IJRITCC | May 2014, Available @ http://www.ijritcc.org

_______________________________________________________________________________________

International Journal on Recent and Innovation Trends in Computing and Communication

Volume: 2 Issue: 5

ISSN: 2321-8169
1081 1085

_______________________________________________________________________________________________

Figure 3.12: Lesion Mask Interface.

Figure 3.9: Geometrical Centre (Binary image)
Next step is the detection process where a graphical user
interface has been created for the detection of the cancer
cells. The figure of GUI is shown below:

Figure 3.13: Detection Interface.

The risk factor found in the lesion is 0.76 i.e. High Risk.
That means the lesion is malignant i.e. Cancerous.
Figure 3.10: Computer aided Diagnosis System

Some sample images have been taken and then analysed for
the detection purpose through the graphical user interface.

Then, the image will be loaded in the interface. Then, the

mask of the lesion will be loaded. And, finally it can be seen
that after analyzing the processes, the lesion malignancy
probability will be shown. The respective figures are shown
below.

Figure 3.14: 1st Image

Figure 3.11: Open Image Interface

Figure 3.15: 2nd Image
1084
IJRITCC | May 2014, Available @ http://www.ijritcc.org

_______________________________________________________________________________________

International Journal on Recent and Innovation Trends in Computing and Communication

Volume: 2 Issue: 5

ISSN: 2321-8169
1081 1085

_______________________________________________________________________________________________

[3]

Figure 3.16: 3rd Image

st

nd

[4]
rd

Samples

1
Image

2 Image

3 Image

Lesion
Probability

0.29254

0.43497

0.39468
[5]

Factors

Benign

Malignant

Suspicious

Table 3.1 Lesion Risk Factors

Thus it is conducted that the detection process of cancerous
cells in skin lesion is carried out.
IV. CONCLUSION
The different methods have been discussed to find the
cancer cells in the skin lesions. The different methods
implemented are image acquisition, segmentation, preprocessing, feature extraction and detection methods. Here
we use different image feature extraction through image
processing methods and they are border, colour, entropy,
compactness, radial variance of the mask, coarseness. Based
on these features, the risk probability factor of the lesion is
shown with the help of computer aided diagnosis system.

[1]

[2]

V. REFERANCES
Yana Goncharova, Enas A. S. Attia, Khawla Souid, and
Inna V. Vasilenko, Dermoscopic Features of Facial
Pigmented Skin Lesions, Hindawi Publishing
Corporation, Dermatology, pp. 1-7, 2013.
Maryam Sadeghi, Timk. Lee, David Mclean, Harvey
Lui, and M. Stella Atkins, Detection and Analysis of

[6]

[7]

[8]

Irregular Streaks in Dermoscopic Images of Skin

Lesions, IEEE Transactions on Medical Imaging, Vol.
32, No. 5, May 2013.
Iris Zalaudek,
M.D, Aimilios Lallas,
Elvira
Moscarella, Caterina Longo, H. Peter Soyer, Giuseppe
Argenziano, The dermatologists applications
traditional and new applications of dermoscopy,
Dermatol Pract Concept, vol.3, no. 2, pp. 67-71, April
30, 2013.
Cheng Lu, Muhammad Mahmood, Naresh Jha, and
Mrinal Mandal, Automated Segmentation of the
Melanocytes in Skin Histopathological Images, IEEE
JOURNAL OF BIOMEDICAL AND HEALTH
INFORMATICS, VOL. 17, NO. 2, pp. 284-296,
MARCH 2013.
[5] Catarina Barata, Margarida Ruela, Mariana
Francisco, Teresa Mendona, and Jorge S. Marques,
Two Systems for the Detection of Melanomas in
Dermoscopy Images Using Texture and Color
Features, IEEE SYSTEMS JOURNAL, pp.1-15,
2013.
Gianluca Sforza, Giovanna Castellano, Sai Krishna
Arika, Robert W. LeAnder, R. Joe Stanley, William V.
Stoecker, and Jason R. Hagerty, Using Adaptive
Thresholding and Skewness Correction to Detect Gray
Areas in Melanoma In Situ Images, IEEE
TRANSACTIONS ON INSTRUMENTATION AND
MEASUREMENT, VOL. 61, NO. 7, pp. 1829-1847,
JULY 2012.
Diego Caratelli, Alessandro Massaro, Roberto
Cingolani, and Alexander G. Yarovoy, Accurate TimeDomain Modelling of Reconfigurable Antenna Sensors
for Non-Invasive Melanoma Skin Cancer Detection,
IEEE SENSORS JOURNAL, VOL. 12, NO. 3, pp. 635645, MARCH 2012.
Kouhei Shimizu, Hitoshi Iyatomi, Kerri-Ann Norton,
M.Emre Celebi, Extension of Automated Melanoma
Screening for Non-Melanocytic Skin Lesions, 19th
International Conference on Mechatronics and Machine
Vision in Practice, pp. 16-19, 2012.

1085
IJRITCC | May 2014, Available @ http://www.ijritcc.org

_______________________________________________________________________________________